First ControlNet attempt

app.py

@@ -1,25 +1,42 @@
+import cv2
 import gradio as gr
+import numpy as np
 import torch
-from diffusers import DiffusionPipeline
+
+from diffusers import DiffusionPipeline, StableDiffusionControlNetPipeline
+from diffusers.utils import load_image
+from PIL import Image
 
 device = "cuda" if torch.cuda.is_available() else "cpu"
 
+image = load_image("https://hf.co/datasets/huggingface/documentation-images/resolve/main/diffusers/input_image_vermeer.png")
+image = np.array(image)
+
+low_threshold = 100
+high_threshold = 200
+
+image = cv2.Canny(image, low_threshold, high_threshold)
+image = image[:, :, None]
+image = np.concatenate([image, image, image], axis=2)
+canny_image = Image.fromarray(image)
+
 def generate(
     prompt, negative_prompt, num_inference_steps, width, height, guidance_scale, seed
 ):
-    pipeline = DiffusionPipeline.from_pretrained("Lykon/DreamShaper")
-    pipeline = pipeline.to(device)
-    generator = torch.Generator(device=device).manual_seed(seed)
-
-    return pipeline(
-        prompt=prompt,
-        negative_prompt=negative_prompt,
-        num_inference_steps=num_inference_steps,
-        width=width,
-        height=height,
-        guidance_scale=guidance_scale,
-        generator=generator,
-    ).images[0]
+    return canny_image
+    # pipeline = DiffusionPipeline.from_pretrained("Lykon/DreamShaper")
+    # pipeline = pipeline.to(device)
+    # generator = torch.Generator(device=device).manual_seed(seed)
+
+    # return pipeline(
+    #     prompt=prompt,
+    #     negative_prompt=negative_prompt,
+    #     num_inference_steps=num_inference_steps,
+    #     width=width,
+    #     height=height,
+    #     guidance_scale=guidance_scale,
+    #     generator=generator,
+    # ).images[0]
 
 iface = gr.Interface(
     fn=generate,

requirements.txt

@@ -1,4 +1,5 @@
 accelerate
 diffusers
+opencv-contrib-python
 torch
 transformers